Connectedness of Commodity Markets: A Dynamic Study on the Effects of Market Crashes and Sentiments

In the first chapter of this thesis, we aim to provide a comprehensive overview of the interconnectedness of commodity markets and to conduct a bibliometric analysis of this topic from 1990 to 2021. To this end, this chapter critically and selectively provide the knowledge map of the connectedness of commodity markets based on the scientific articles published on the Web of Science (WoS). In doing this, we group the literature survey based on notable commodity markets and provide an overview of the empirical literature based on single- and cross-commodity markets. The key finding of the literature survey is that there is connectedness within and across commodity markets, with evidence of time variations triggered largely by global financial crises. In addition, from 144 articles over the last two decades, significant conceptual clusters and networks arise, which suggest a close density of networks in terms of the keyword clusters, keyword plus co-occurrences, country collaborations, and journal co citations. Furthermore, there are significant conceptual clusters that cover the association of connectedness type, commodity market, type of statistical analysis, association of major energy shocks, futures market, co-movement, and association of transmission in stock and gold markets. Our analysis, therefore, suggests, among other things, the need for future research to analyse the pricing of pollution credits as the newest commodity market. This finding is useful for economic actors, investors, and policymakers have a better understanding of the dynamic behaviour of commodity prices. In the second chapter of this thesis, we examine the impact of the COVID-19 pandemic on major agricultural commodity prices (cattle, cocoa, coffee, corn, cotton, hog, rice, soya oil, soybeans, soybean meal, sugar and wheat) using daily data from 1 January 2016 to 25 February 2022. We measured COVID-19 effect using a news-based sentiment index. A robust nonparametric Granger causality-in-quantiles test is used to test the effect of the COVID-19 sentiment on agricultural commodity prices and price volatility. We find significant Granger causality from the news-based COVID-19 sentiment to mean of the agricultural commodity prices in the lower and upper ranges of the quantiles. Moreover, findings show that the COVID-19 sentiment is also causal for variance of agricultural commodity prices, but only above the quantile ranges above the first quarter. Thus, COVID-19 is causal for large volatility changes in agricultural commodity prices. Accordingly, the extremely negative sentiment associated with COVID-19 has not only caused a price crash in agricultural markets, but also significantly increased market risk. Policymakers should be cautious of the potential risks and vulnerabilities that agricultural commodities may face in the event of extreme circumstances, as well as the potential consequences for producers and consumers throughout the economy.